Cool. I forked the repo and made some commits to reflect the current state of the site. One thing I noticed is that, instead of pulling directly from the database, much of the content is pulled from a “cache” directory that seems to be composed of html files generated by some PHP scripts in “cache-generator” directory. For some reason, I couldn’t get all of these files (some 4000!) to be re-generated this way, so I moved this cache-generator directory into “scripts” (was in the root directory) so it could be studied.
It’s probably worth thinking about what a re-architected version of the site would look like. Thoughts, @mike & @swmerrill?
Odd that the PHP would generate HTML files and put them in a cache directory!
If we’re thinking of a re-architected site, I’m curious about what you see it looking like. My primary interest is in having a database in which to look up product information using some kind of API. That presupposes that the database is kept up-to-date. I’m wondering what else people might want.
The first thing I’d like to do, though, is figure out why the Android app (and possibly the iOS app) cannot retrieve data now that the site is back online. When I get a minute I’ll check whether the code for those is in the repository.
@swmerrill how is it going? I know we got site back online but we have subsequently had issues on server s033 (where I think it was living). Can we migrate this site across to s110?
Hello All, I am the founder of a Tech company that is starting to catalog global TV commercials, and the products that appear in them. Product details, Pictures, Associated Brands, Industry Metadata, etc etc… as well as the TV commercials production information and cast. Demo is at www.anuncio.media if anyone interested. The product information will be open source, and is being populated by a team at the moment. I am looking for contributors, or to partnerships and act as a data source for product-open-data… Please get in contact me - stephen@anuncio.guru if interested.
I have just discover Open Product Data, It’s quite cool and neccessary, the
thing is that I have been working in a quite similiar project WikiObject https://meta.wikimedia.org/wiki/WikiObject
The idea to “resurrect product-open-data” could be use wikidata as infrastructure for storing product data information, and use wikibase or another client to show the information in a user friendly way.
I am not a wikimedia reponsible, developer or whatever so I only had feedback from wikimedia community through forums (see https://meta.wikimedia.org/wiki/WikiObject#Discussion). I think that begin conversations with/between OKFN/Wikimedia stakeholders should be good since this idea have too much implications to start without join forces.
Note: I’m unable to add links to my posts, so I’ve had to obscure the URLs
In case anyone finds their way here while searching for the [https]://product-open-data.com data, I’ve found some copies from 2013 and archived them on archive[dot]org
Here’s a gtin database dump (postgres dump, no images) from 2016: [https]://archive.org/details/opd_pg_2016.04.08_01
Here is an image dump from 2013 (2.5G zip file): [https]://archive.org/details/pod_pictures_gtin_2013.08.29_01
As of 01 April 2020, the images are still stored in an S3 bucket [http]://product.okfn.org.s3.amazonaws.com/images/
The GTIN images have paths like this: [http]://product.okfn.org.s3.amazonaws.com/images/gtin/gtin-528/5283001118546.jpg
The code for product-open-data[dot]org (as of 01 April 2020) is here: [https]://github.com/okfn/labs-opd
Per the GTIN product page template [https]://github.com/okfn/labs-opd/blob/3386d026ab66aa2734409034df4afaabb1a8309c/browser/templates/browser/gtin.html, the URLs for the GTIN image S3 can be constructed like this:
Not sure if this conversation is ongoing, but I’m interested in connecting with other people with an interest in making GTIN data available.
I’ve taken a snapshot of the data from product-open-data and made it available here: https:///www.brocade.io
The plan I’ve in mind is to work out protocols for sharing data between these kinds of systems. I’ve a vague idea that something like this exists already within GEPIR’s ecosystem, but I’m still learning about how all this works.
Anyway, just reaching out. Happy to help support product-open-data any way I can.